A Computational Implementation of a Human Attention Guiding Mechanism in MIDAS v5

摘要

In complex human-machine systems, the human operator is often required to intervene to detect and solve problems. Given this increased reliance on the human in these critical human-machine systems, there is an increasing need to validly predict how operators allocate their visual attention. This paper describes the information-seeking (attention-guiding) model within the Man-machine Integration Design and Analysis System (MIDAS) v5 software - a predictive model that uses the Salience, Effort, Expectancy and Value (SEEV) of an area of interest to guide a person's attention. The paper highlights the differences between using a probabilistic fixation approach and the SEEV approach in MIDAS to drive attention.